CN109426978A - Method and apparatus for generating information - Google Patents

Method and apparatus for generating information Download PDF

Info

Publication number
CN109426978A
CN109426978A CN201710758012.4A CN201710758012A CN109426978A CN 109426978 A CN109426978 A CN 109426978A CN 201710758012 A CN201710758012 A CN 201710758012A CN 109426978 A CN109426978 A CN 109426978A
Authority
CN
China
Prior art keywords
feedback information
information
end article
mass fraction
negative
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201710758012.4A
Other languages
Chinese (zh)
Inventor
向彪
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Jingdong Century Trading Co Ltd
Beijing Jingdong Shangke Information Technology Co Ltd
Original Assignee
Beijing Jingdong Century Trading Co Ltd
Beijing Jingdong Shangke Information Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Jingdong Century Trading Co Ltd, Beijing Jingdong Shangke Information Technology Co Ltd filed Critical Beijing Jingdong Century Trading Co Ltd
Priority to CN201710758012.4A priority Critical patent/CN109426978A/en
Publication of CN109426978A publication Critical patent/CN109426978A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/30—Semantic analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/01—Customer relationship services
    • G06Q30/015—Providing customer assistance, e.g. assisting a customer within a business location or via helpdesk
    • G06Q30/016—After-sales

Abstract

The embodiment of the present application discloses the method and apparatus for generating information.One specific embodiment of this method includes: to obtain user in prefixed time interval to be directed at least one feedback information that end article is submitted；Semantic analysis is carried out at least one feedback information, determines whether every feedback information at least one feedback information is the negative feedback information for being used to indicate end article there are quality problems；Determine the quantity of feedback information negative at least one feedback information；According to the sum of feedback information in the quantity of negative feedback information and at least one feedback information, for characterizing end article, there are the mass fractions of the probability of quality problems for generation.This embodiment offers a kind of commercial quality monitoring mechanism based on feedback information, enriches commercial quality monitoring method.

Description

Method and apparatus for generating information
Technical field
This application involves field of computer technology, and in particular to Internet technical field, more particularly, to generation information Method and apparatus.
Background technique
With the continuous development of e-commerce, consumer has higher pursuit to the quality of commodity.Currently, commercial quality Monitoring is mainly carried out by traditional means such as sampling inspection.
However, since electric business platform is commodity circulation, a transaction platform.It, which has, is much different from traditional mode of production manufacture The characteristics of enterprise.For example, the ranks such as type and quantity of commodity up to ten million, hundred million, 1,000,000,000, and merchandise resources are extremely complex, together A kind of commodity may be from different suppliers or different production batch, in addition, some third party businessman is passed through The commodity of battalion are directly delivered by businessman, and electric business platform is caused to become more to be stranded for the control of merchandise resources, batch and quality It is difficult.For the These characteristics of electric business platform, the purpose of quality monitoring is unable to reach using traditional Quality Control means.
Summary of the invention
The purpose of the embodiment of the present application is to propose a kind of improved method and apparatus for generating information, come solve with The technical issues of upper background technology part is mentioned.
In a first aspect, the embodiment of the present application provides a kind of method for generating information, this method comprises: obtaining default User is directed at least one feedback information that end article is submitted in time interval；Semantic point is carried out at least one feedback information Analysis determines whether every feedback information at least one feedback information is to be used to indicate end article bearing there are quality problems The feedback information in face；Determine the quantity of feedback information negative at least one feedback information；According to negative feedback information The sum of feedback information in quantity and at least one feedback information is generated for characterizing probability of the end article there are quality problems Mass fraction.
In the present embodiment, semantic analysis is carried out at least one feedback information, determined at least one feedback information Whether every feedback information is the negative feedback information for being used to indicate end article there are quality problems, comprising: at least one Every feedback information in feedback information is pre-processed, and keyword sequence corresponding with every feedback information is generated；According to Obtained keyword sequence determines term vector matrix corresponding with every feedback information；Identified term vector matrix is led Enter semantics recognition model trained in advance, obtains the semanteme of every feedback information, semantics recognition model is for characterizing term vector square The semantic corresponding relationship of battle array and feedback information, the semanteme of feedback information include negative and non-negative.
In the present embodiment, method further include: the feature of mass fraction is determined according at least one in following item of information Information: the sum of feedback information at least one feedback information, the number of users for submitting at least one feedback information, pre-set The attribute value of end article.
In the present embodiment, according in the quantity of negative feedback information and at least one feedback information feedback information it is total Number, for characterizing end article, there are the mass fractions of the probability of quality problems for generation, comprising: determines negative feedback information The ratio of the sum of feedback information in quantity and at least one feedback information；According to ratio and characteristic information, mass fraction is determined.
In the present embodiment, according in the quantity of negative feedback information and at least one feedback information feedback information it is total Number, for characterizing end article, there are the mass fractions of the probability of quality problems for generation, comprising: according at least one feedback information In every feedback information submission time, at least one feedback information is divided into preset number feedback information set；Really The quantity for determining the negative feedback information that each feedback information set includes in preset number feedback information set and includes The ratio of the sum of feedback information；It obtains and is directed to each pre-set weight of feedback information set；According to each feedback information Gather corresponding weight and ratio, determines mass fraction.
In the present embodiment, method further include: determine the quotient that every feedback information is targeted at least one feedback information The time difference between time that the delivery availability and feedback information of product are submitted；Determine that end article is flat according to the identified time difference Feed back lag time.
In the present embodiment, it obtains user in prefixed time interval and is directed at least one feedback letter that end article is submitted Breath, comprising: inquire submission time within a preset time interval in historical feedback information aggregate, and believe with the mark of end article Cease matched feedback information；Obtain in the feedback information that inquires, with the associated order information of feedback information inquired and/or The user information not feedback information in pre-set blacklist.
In the present embodiment, method further include: be greater than preset first threshold in response to mass fraction generated, generate It is used to indicate prompt information of the end article there may be quality problems.
In the present embodiment, method further include: determine the mass fraction of mass fraction generated Yu a upper time interval Between difference；It is greater than preset second threshold in response to difference, generation is used to indicate end article, and there may be quality problems Prompt information.
In the present embodiment, method further include: mass fraction and institute in response to preset number time interval before The mass fraction of generation successively decreases, and generation is used to indicate prompt information of the end article there may be quality problems.
In the present embodiment, method further include: to removing target in mass fraction generated and the affiliated classification of end article The mass fraction of other commodity outside commodity is ranked up；According to ranking results determine whether generate be used to indicate end article can There can be the prompt information of quality problems.
Second aspect, the embodiment of the present application provide it is a kind of for generating the device of information, the device include: obtain it is single Member is directed at least one feedback information that end article is submitted for obtaining user in prefixed time interval；First determination unit, For at least one feedback information carry out semantic analysis, determine every feedback information at least one feedback information whether be It is used to indicate negative feedback information of the end article there are quality problems；Second determination unit, for determining that at least one is anti- The quantity of negative feedback information in feedforward information；First generation unit, for according to the quantity of negative feedback information and at least The sum of feedback information in one feedback information, for characterizing end article, there are the quality of the probability of quality problems point for generation Number.
In the present embodiment, the first determination unit, comprising: subelement is generated, for at least one feedback information Every feedback information is pre-processed, and keyword sequence corresponding with every feedback information is generated；First determines subelement, is used for According to obtained keyword sequence, term vector matrix corresponding with every feedback information is determined；Subelement is imported, is used for institute Determining term vector matrix imports semantics recognition model trained in advance, obtains the semanteme of every feedback information, semantics recognition mould Type is used to characterize the semantic corresponding relationship of term vector matrix and feedback information, and the semanteme of feedback information includes negative and non-negative Face.
In the present embodiment, device further include: third determination unit, for true according at least one in following item of information Determine the characteristic information of mass fraction: the sum of feedback information at least one feedback information submits at least one feedback information The attribute value of number of users, pre-set end article.
In the present embodiment, the first generation unit, comprising: second determines subelement, for determining negative feedback information Quantity and at least one feedback information in feedback information sum ratio；Third determines subelement, for according to ratio and Characteristic information determines mass fraction.
In the present embodiment, the first generation unit, comprising: subelement is divided, for according at least one feedback information Every feedback information submission time, at least one feedback information is divided into preset number feedback information set；4th Subelement is determined, for determining each feedback information set includes in preset number feedback information set negative feedback letter The quantity of breath, the ratio with the sum for the feedback information for including；First obtains subelement, is directed to each feedback information for obtaining Gather pre-set weight；5th determines subelement, is used for according to the corresponding weight of each feedback information set and ratio, really Determine mass fraction.
In the present embodiment, device further include: the 4th determination unit, for determine at least one feedback information every it is anti- The time difference between time that the delivery availability and feedback information of the targeted commodity of feedforward information are submitted；5th determination unit is used In determining end article average feedback lag time according to the identified time difference.
In the present embodiment, acquiring unit, comprising: inquiry subelement is mentioned for inquiring in historical feedback information aggregate Hand over the time within a preset time interval, and the feedback information with the identification information match of end article；Second obtains subelement, uses In obtaining in the feedback information that inquires, with the associated order information of feedback information and/or user information that inquire not pre- The feedback information in blacklist being first arranged.
In the present embodiment, device further include: the second generation unit, it is pre- for being greater than in response to mass fraction generated If first threshold, generation be used to indicate prompt information of the end article there may be quality problems.
In the present embodiment, device further include: the 6th determination unit, for determining mass fraction generated and upper a period of time Between difference between the mass fraction that is spaced；Third generation unit is generated for being greater than preset second threshold in response to difference It is used to indicate prompt information of the end article there may be quality problems.
In the present embodiment, device further include: the 4th generation unit, in response between the preset number time before Every mass fraction successively decrease with mass fraction generated, generation be used to indicate prompt of the end article there may be quality problems Information.
In the present embodiment, device further include: sequencing unit, for belonging to mass fraction generated and end article The mass fraction of other commodity in classification in addition to end article is ranked up；5th generation unit, for according to ranking results Determine whether to generate the prompt information for being used to indicate end article there may be quality problems.
The third aspect, the embodiment of the present application provide a kind of equipment, comprising: one or more processors；Storage device is used In storing one or more programs, when said one or multiple programs are executed by said one or multiple processors, so that above-mentioned One or more processors realize such as the above-mentioned method of first aspect.
Fourth aspect, the embodiment of the present application provide a kind of computer readable storage medium, are stored thereon with computer journey Sequence, which is characterized in that such as first aspect above-mentioned method is realized when the program is executed by processor.
Method and apparatus provided by the embodiments of the present application for generating information, by obtaining user in prefixed time interval For at least one feedback information that end article is submitted, semantic analysis then is carried out at least one feedback information, is determined extremely Whether every feedback information in a few feedback information is the negative feedback for being used to indicate end article there are quality problems Information, and determine the quantity of feedback information negative at least one feedback information, finally according to the number of negative feedback information The sum of feedback information in amount and at least one feedback information, for characterizing end article, there are the probability of quality problems for generation Mass fraction enriches commercial quality monitoring method to provide a kind of commercial quality monitoring mechanism based on feedback information.
Detailed description of the invention
By reading a detailed description of non-restrictive embodiments in the light of the attached drawings below, the application's is other Feature, objects and advantages will become more apparent upon:
Fig. 1 is that this application can be applied to exemplary system architecture figures therein；
Fig. 2 is the flow chart according to one embodiment of the method for generating information of the application；
Fig. 3 is a schematic diagram according to the application scenarios of the method for generating information of the application；
Fig. 4 is the flow chart according to another embodiment of the method for generating information of the application；
Fig. 5 is the structural schematic diagram according to one embodiment of the device for generating information of the application；
Fig. 6 is adapted for the structural schematic diagram for the computer system for realizing the server of the embodiment of the present application.
Specific embodiment
The application is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched The specific embodiment stated is used only for explaining related invention, rather than the restriction to the invention.It also should be noted that in order to Convenient for description, part relevant to related invention is illustrated only in attached drawing.
It should be noted that in the absence of conflict, the features in the embodiments and the embodiments of the present application can phase Mutually combination.The application is described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
Fig. 1 is shown can be using the method for generating information of the application or the implementation of the device for generating information The exemplary system architecture 100 of example.
As shown in Figure 1, system architecture 100 may include terminal device 101,102,103, network 104 and server 105, 106.Network 104 between terminal device 101,102,103 and server 105,106 to provide the medium of communication link.Net Network 104 may include various connection types, such as wired, wireless communication link or fiber optic cables etc..
User 110 can be used terminal device 101,102,103 and be interacted by network 104 with server 105,106, to connect Receive or send data etc..Various applications, such as the application of shopping class, map class can be installed on terminal device 101,102,103 Using, payment class application, social category application, web browser applications, search engine class apply, mobile phone assistant's class apply etc..
Terminal device 101,102,103 can be with display screen and the various electronics of data communication function supported to set It is standby, including but not limited to smart phone, tablet computer, E-book reader, MP3 player (Moving Picture Experts Group Audio Layer III, dynamic image expert's compression standard audio level 3), MP4 (Moving Picture Experts Group Audio Layer IV, dynamic image expert's compression standard audio level 4) player, knee Mo(u)ld top half portable computer and desktop computer etc..User can be uploaded by terminal device 101,102,103 to server anti- The data such as feedforward information.
Server 105,106 can be to provide the server of various services, such as to pacifying on terminal device 101,102,103 The application of dress provides the background server supported, user is directed to target in server 105,106 available prefixed time intervals At least one feedback information that commodity are submitted then carries out semantic analysis at least one feedback information, determines that at least one is anti- Whether every feedback information in feedforward information is the negative feedback information for being used to indicate end article there are quality problems, and really The quantity of negative feedback information in fixed at least one feedback information, finally according to the quantity of negative feedback information and at least one The sum of feedback information in feedback information, generates that there are the mass fractions of the probability of quality problems for characterizing end article.
It should be noted that the embodiment of the present application provided by for generate information method can by server 105, 106 execute, and correspondingly, the device for generating information can be set in server 105,106.
It should be understood that the number of terminal device, network and server in Fig. 1 is only schematical.According to realization need It wants, can have any number of terminal device, network and server.
With continued reference to Fig. 2, the process of one embodiment of the method for generating information according to the application is shown 200.The method for being used to generate information, comprising the following steps:
Step 201, it obtains user in prefixed time interval and is directed at least one feedback information that end article is submitted.
In the present embodiment, the method for generating information runs electronic equipment (such as service shown in FIG. 1 thereon Device) at least one feedback information that user in prefixed time interval is directed to end article submission can be obtained first.Preset time Interval can be set according to actual needs, such as can be 5~10 days, when suitably can increase default when feedback information is less Between be spaced.End article can be any required commodity for determining its quality condition, and end article may include one or more Commodity, when end article includes a variety of commodity, end article can be multiple commodity of same commodity classification or same brand.Quotient Category mesh refers to the form that commodity classification is not organized from big to small, such as wine, can use household items -> drinks- > grape wine/white wine, such a three-level classification administrative mechanism.Namely: grape wine and white wine belong to household items (level-one class Mesh) drinks (second level classification) classification in the following, the grape wine of certain specific brand just in grape wine (three-level classification) in the following, certain product The white wine of board just white wine (three-level classification) below.
In the present embodiment, feedback information may include user comment in e-commerce website or other related web sites Information, user replace the information repaired in application record, user and the chat message of customer service, the calling information after sale of user etc..Instead Feedforward information can store in the feedback information table of database, can also be stored in feedback information table corresponding with feedback information The feedback date, the numbers of the targeted commodity or order of feedback information, the type of feedback information (comment on, replace repair, seek advice from, Complain), the mark of feedback user etc..Wherein, the number of commodity can be SKU (Stock Keeping Unit, quantity in stock list Position) or other numbers that can identify commodity, the mark of feedback user can be account, cell-phone number of user etc..With target quotient For product for for certain shoes, feedback information be can be " shoes, which are not worn two days, just to come unglued ", the feedback date can be " 20170208 ", The number of commodity can be " 12345 ", and the number of order can be " 24680 ", and feedback kind can be " comment ", feedback user Mark can be " abc123 ".
In some optional implementations of the present embodiment, obtains user in prefixed time interval and submitted for end article At least one feedback information, comprising: in historical feedback information aggregate inquire submission time within a preset time interval, and with The feedback information of the identification information match of end article；It obtains in the feedback information inquired, is closed with the feedback information inquired The order information and/or user information of the connection not feedback information in pre-set blacklist.
In this implementation, due to may exist in e-commerce website wash sale or malice comment ask Topic, need to exclude the associated feedback information of user or order for being related to problems as interference feedback information, In order to avoid influencing the accuracy for end article quality evaluation.Being related to for the air control system identification of electric business platform being gone out is such The user of problem or the information of order are added in blacklist, for using when subsequent exclusive PCR feedback information.
Step 202, semantic analysis is carried out at least one feedback information, determines that every at least one feedback information is anti- Whether feedforward information is the negative feedback information for being used to indicate end article there are quality problems.
In the present embodiment, above-mentioned electronic equipment can carry out language at least one feedback information obtained in step 201 Justice analysis determines whether every feedback information at least one feedback information is to be used to indicate end article there are quality problems Negative feedback information.Semantic analysis can be with various machine learning methods, excavate general with learning text profound level It reads.For example, can generate model by convolutional neural networks or text subject obtains the theme distribution of text, text subject is generated Model can be LDA (Latent Dirichlet Allocation implies the distribution of Di Li Cray) model, hLDA (hierarchical Latent Dirichlet Allocation is layered implicit Di Li Cray distribution) model or HDP (hierarchical Dirichlet process is layered Di Li Cray process) model.
In the present embodiment, semantic analysis be also possible to the negative and non-negative feedback information that will be marked in advance as Positive negative sample is trained initial model based on positive negative sample to obtain the disaggregated model for classifying to feedback information, classification Model be also possible to technical staff pre-established based on the statistics to a large amount of feedback information and classification results, be stored with it is more The mapping table of the corresponding relationship of a keyword and classification results.
Step 203, the quantity of feedback information negative at least one feedback information is determined.
In the present embodiment, above-mentioned electronic equipment can determine at least one according to the semantic analysis result in step 202 End article is used to indicate in feedback information, and there are the quantity of the negative feedback information of quality problems.
Step 204, raw according to the sum of feedback information in the quantity of negative feedback information and at least one feedback information At for characterizing end article, there are the mass fractions of the probability of quality problems.
In the present embodiment, above-mentioned electronic equipment can be according to the quantity of the negative feedback information determined in step 203 With the sum of feedback information at least one feedback information that obtains in step 201, generate that there are matter for characterizing end article The mass fraction of the probability of amount problem.The ratio of negative feedback information is higher at least one feedback information, and end article is deposited It is possible larger in the probability of quality problems.
In some optional implementations of the present embodiment, method further include: according at least one in following item of information Determine the characteristic information of mass fraction: the sum of feedback information, at least one feedback information of submission at least one feedback information Number of users, pre-set end article attribute value.
In this implementation, characteristic information can serve to indicate that the different degree of mass fraction, credible speed or reference price Value.Attribute value can be the value of reflection commodity significance level, can determine different classifications in conjunction with the demand of merchandise control level The corresponding attribute value of commodity specifically a possibility that quality problems can occur according to commodity and quality problems consequence occur in commodity Severity determines attribute value, for example, the harm that quality problems occurs in food commodity is generally higher than the appearance of shoes cap commodity The harm of quality problems, so the attribute value of food commodity can be greater than the attribute value of shoes cap commodity.
In view of the feedback time of feedback information, the quantity of feedback information, the number of users for submitting feedback information, etc. factors It influences, can individually can then export characteristic information using above-mentioned factor as the characteristic information of mass fraction, it can also basis The sum of feedback information is common in the quantity of characteristic information and negative feedback information, at least one feedback information determines quality point Number.
In some optional implementations of the present embodiment, fed back according to the quantity of negative feedback information and at least one The sum of feedback information in information, for characterizing end article, there are the mass fractions of the probability of quality problems for generation, comprising: really The ratio of the sum of feedback information in the quantity of fixed negative feedback information and at least one feedback information；According to ratio and feature Information determines mass fraction.
In this implementation, when end article includes multiple commodity of same commodity classification, determining mass fraction It can reflect the average quality of the commodity classification.When end article is the same kind of goods, determining mass fraction can reflect this The average quality of commodity.For example, mass fraction can be calculated by the following formula:
Wherein, AQ indicates that the mass fraction of commodity classification or the mass fraction of commodity, A indicate the number of negative feedback information Amount, B indicate the sum of feedback information at least one feedback information.W indicates the characteristic information of mass fraction, can be according at least Number of users, the pre-set end article of the sum of feedback information, at least one feedback information of submission in one feedback information Attribute value etc. determine the characteristic information of mass fraction, for example, mass fraction can be calculated by the following formula:
Wherein, x can be lg (B), the logarithm for being bottom B with 10 be indicated, so only when the sum of feedback information reaches one In the case where fixed number amount, average quality ability confidence level with higher.The weight, which also can according to need, uses other weighting calculation instead Method.Same x can also according to submit the number of users of at least one feedback information, attribute value of pre-set end article etc. because Element determines.
In some optional implementations of the present embodiment, fed back according to the quantity of negative feedback information and at least one The sum of feedback information in information, for characterizing end article, there are the mass fractions of the probability of quality problems for generation, comprising: root According to the submission time of every feedback information at least one feedback information, at least one feedback information is divided into preset number A feedback information set；Determine the negative feedback letter that each feedback information set includes in preset number feedback information set The quantity of breath, the ratio with the sum for the feedback information for including；It obtains and is directed to each pre-set weight of feedback information set； According to the corresponding weight of each feedback information set and ratio, mass fraction is determined.
In this implementation, the rule for dividing feedback information can be configured according to actual needs, for example, feedback letter Breath obtains in 5 days in the past, then can daily divide.The weight of each feedback information set can according to actual needs into Row setting, for example, can determine the weight of feedback information set by following formula:
Wherein,Indicate cooling ratio,It can be configured according to actual needs, for example, can be set to 0.2.WiTable The weight for showing i-th of feedback information set, by taking feedback information is to obtain in 5 days in the past as an example, the determining end article at the t days Mass fraction when, the feedback information set marked off can respectively include the feedback information got for t-i days, and i can take 1 ~5.
In this implementation, the quantity for the negative feedback information that can include according to feedback information set and includes Feedback information sum ratio, the mass fraction of each feedback information set is determined, for example, following formula meter can be passed through Calculate the mass fraction of single feedback information set:
Wherein, QiIndicate the mass fraction of i-th of feedback information set, C indicates negative anti-in the feedback information set The quantity of feedforward information, D indicate the sum of feedback information in the feedback information set.W indicates the quality point of the feedback information set Several characteristic informations, can according to the sum of feedback information in the feedback information set, submit in the feedback information set and feed back The number of users of information, attribute value of pre-set end article etc. determine, for example, the feedback can be calculated by the following formula The mass fraction of information aggregate:
Wherein, y can be log2D indicates the logarithm for being bottom D with 2, so only when feedback letter in the feedback information set In the case that the sum of breath reaches certain amount, the mass fraction ability confidence level with higher of the feedback information set.Equally Also can according to need and use other weighting algorithm instead, according to submit the feedback information set in feedback information number of users, in advance Attribute value of the end article of setting etc. determines y.
In this implementation, according to the corresponding weight of each feedback information set and ratio, mass fraction is determined, it can be with It is that the mass fraction of end article is determined according to the corresponding weight of each feedback information set and mass fraction, for example, can lead to Cross the mass fraction that following formula calculates end article:
Wherein, Q indicates that the mass fraction of end article, m indicate the number of feedback information set.
In some optional implementations of the present embodiment, method further include: determine at least one feedback information every The time difference between time that the delivery availability and feedback information of the targeted commodity of feedback information are submitted；According to it is identified when Between difference determine end article average feedback lag time.
In this implementation, the delivery availability of the targeted commodity of feedback information can be obtained from order information database Take, the time that can then submit the delivery availability of the targeted commodity of identified every feedback information and feedback information it Between the arithmetic average of time difference it is true that other statistical methods also can be used as end article average feedback lag time Make end article average feedback lag time.End article average feedback lag time can be used for commodity tracing, i.e., when out When existing abnormal quality phenomenon, it counter can release that there may be the information such as the commerical batches of problem.
In the present embodiment, it generates for characterizing end article there are after the mass fraction of the probability of quality problems, it can To show mass fraction, or the terminal push mass fraction used to the personnel of responsible quality monitoring, it can also be according to quality point Number to carry out quality-monitoring to commodity classification or single commodity, with lasting promotion quality level, finds abnormal quality in time.
In some optional implementations of the present embodiment, method further include: be greater than in response to mass fraction generated Preset first threshold, generation are used to indicate prompt information of the end article there may be quality problems.
In this implementation, different first thresholds can be set for inhomogeneity purpose end article, for example, being directed to The first threshold of food setting should be greater than the first threshold for shoes and hats class assignment.First threshold can also be according to can be with root It determines according to for the preset long term quality target of end article, specifically can be split out more according to preset long term object A node destination is come pair using modes such as linear regressions according to the year-on-year and current ring ratio of current quality trends and last year The quality of the following certain time is predicted.If it was found that generation is used to indicate target when prediction target could possibly be higher than predetermined value There may be the prompt informations of quality problems for commodity, the subsequent actions such as anomaly analysis and investigation can also be triggered, for example, can root According to Quality Control process, progress data analysis is showed anomalous mass, and targetedly formulates measures to rectify and reform.
In some optional implementations of the present embodiment, method further include: determine mass fraction generated and upper one Difference between the mass fraction of time interval；It is greater than preset second threshold in response to difference, generation is used to indicate target quotient There may be the prompt informations of quality problems for product.
In this implementation, fluctuation monitoring is carried out to mass fraction, specifically can be each time interval of monitoring The fluctuating range and ratio of mass fraction, for example, the ring ratio for the mass fraction that the mass fraction of same day determination and the previous day determine. If fluctuating range or ratio within range of set value (such as one thousandth), do not trigger task subsequent action, otherwise will give birth to At prompt information of the end article there may be quality problems is used to indicate, the subsequent actions such as anomaly analysis and investigation are triggered.
In some optional implementations of the present embodiment, method further include: in response to the preset number time before The mass fraction at interval successively decreases with mass fraction generated, and generation is used to indicate end article mentioning there may be quality problems Show information.
In this implementation, Data Trend Monitor has been carried out to mass fraction, can specifically monitor continuous several time intervals The variation tendency of interior mass fraction, to determine that the mass change of end article belongs to random short-term normal fluctuation or tendency Variation.It can (abscissa and Time of Day, ordinate be that the commodity classification that be averaged is averaged by the daily tendency chart of rendering quality score Quality point), and certain exception-triggered condition is set according to being actually configured, for example, it is higher than within continuous 5 days a certain threshold value, or The consistent situation that degenerates is presented within person continuous 5 days, reaches exception-triggered condition, then generates and be used to indicate end article there may be matter The prompt information of amount problem.
In some optional implementations of the present embodiment, method further include: to mass fraction generated and target quotient The mass fraction of other commodity in the affiliated classification of product in addition to end article is ranked up；Determine whether to generate according to ranking results It is used to indicate prompt information of the end article there may be quality problems.
In this implementation, if the mass fraction of end article is forward, for example, preceding 1% (specific value can be according to reality Border needs to be configured), then the prompt information for being used to indicate end article there may be quality problems can be generated.Due to difference There is very big differences usually in quantized values for commercial quality between classification, for example use identical statistical method, family The mass fraction of the electric class mass fraction of commodity and foodstuff commodity now now there is very big difference, this be usually by It is codetermined in the Various Complexes factor such as industry itself, crowd, environment, so the mass fraction to commodity in same classification carries out The result of sequence has more reference value.
With continued reference to the signal that Fig. 3, Fig. 3 are according to the application scenarios of the method for generating information of the present embodiment Figure.In the application scenarios of Fig. 3, server 301 obtains user in prefixed time interval first and passes through terminal 302,303 needle of terminal Semantic analysis is carried out to the feedback information 304 that certain shoes is submitted, and to feedback information, defines wherein negative feedback letter Breath 305, finally according to the sum of the quantity of negative feedback information 305 and feedback information 304, generates for characterizing target quotient There are the mass fractions 306 of the probability of quality problems for product.
The method provided by the above embodiment of the application passes through user in acquisition prefixed time interval first and is directed to target quotient At least one feedback information that product are submitted then carries out semantic analysis at least one feedback information, according to negative feedback letter The sum of feedback information in the quantity of breath and at least one feedback information, for characterizing end article, there are quality problems for generation The mass fraction of probability enriches commercial quality prison to provide a kind of commercial quality monitoring mechanism based on feedback information Prosecutor method.
With further reference to Fig. 4, it illustrates the processes 400 of another embodiment of the method for generating information.The use In the process 400 for the method for generating information, comprising the following steps:
Step 401, it obtains user in prefixed time interval and is directed at least one feedback information that end article is submitted.
In the present embodiment, the method for generating information runs electronic equipment (such as service shown in FIG. 1 thereon Device) at least one feedback information that user in prefixed time interval is directed to end article submission can be obtained first.
Step 402, every feedback information at least one feedback information is pre-processed, is generated and every feedback letter Cease corresponding keyword sequence.
In the present embodiment, above-mentioned electronic equipment can be to every at least one feedback information obtained in step 401 Feedback information is pre-processed, and keyword sequence corresponding with every feedback information is generated.Pretreatment may include participle, delete Except stop words etc. operates, due to containing many terms lack of standardization, punctuation mark, invalid comment etc. in certain customers' feedback content, So needing to pre-process further to standardize the text of feedback information, then the text of feedback information is divided Word, participle are after the completion again removed such as stop words, symbol.
Step 403, according to obtained keyword sequence, term vector matrix corresponding with every feedback information is determined.
In the present embodiment, above-mentioned electronic equipment keyword sequence according to obtained in step 402, using disclosed Term vector library or the term vector library constructed by machine learning method, the corresponding vector representation method of word after searching participle, To determine term vector matrix corresponding with every feedback information.
Step 404, identified term vector matrix is imported to semantics recognition model trained in advance, obtains every feedback letter The semanteme of breath.
In the present embodiment, the term vector matrix determined in step 403 can be imported training in advance by above-mentioned electronic equipment Semantics recognition model, obtain the semanteme of every feedback information.Semantics recognition model is for characterizing term vector matrix and feedback letter The semantic corresponding relationship of breath, the semanteme of feedback information include negative and non-negative.
In the present embodiment, the training sample of semantics recognition model can from existing feedback information picking part text It is marked, marks whether it is negative feedback information, and pre-process to it, to be used for model training.As an example, choosing The sample text taken can be divided into three classes: commenting on, replace and repair, seek advice from after sale and complain.Every class text is according to the one of all commodity Grade classification (such as household items), it is each to extract 50,000, wherein negative feedback information and the text scale of non-negative feedback information are about For 1:4.Can be using 80% sample data as training data, the data of remainder 20% use engineering as test data Algorithm is practised, the initial semantics recognition model of training obtains semantics recognition model, with the accurate rate of testing algorithm and can then call together The rate of returning.Initial semantics recognition model can be a CNN (Convolutional Neural Network, convolutional Neural net Network) model.
Step 405, the quantity of feedback information negative at least one feedback information is determined.
In the present embodiment, the semanteme of above-mentioned electronic equipment every feedback information according to obtained in step 404, really Being used to indicate end article in fixed at least one feedback information, there are the quantity of the negative feedback information of quality problems.
Step 406, raw according to the sum of feedback information in the quantity of negative feedback information and at least one feedback information At for characterizing end article, there are the mass fractions of the probability of quality problems.
In the present embodiment, above-mentioned electronic equipment can be according to the quantity of the negative feedback information determined in step 405 With the sum of feedback information at least one feedback information that obtains in step 401, generate that there are matter for characterizing end article The mass fraction of the probability of amount problem.
In the present embodiment, step 401, step 405, the operation of step 406 and step 201, step 203, step 204 Operate essentially identical, details are not described herein.
Figure 4, it is seen that the method for generating information compared with the corresponding embodiment of Fig. 2, in the present embodiment Process 400 in semantics recognition carried out to pretreated feedback information by semantics recognition model trained in advance, as a result, this In the scheme of embodiment description more preferably to the recognition effect of feedback information, to improve the accuracy of the mass fraction of generation.
With further reference to Fig. 5, as the realization to method shown in above-mentioned each figure, this application provides one kind for generating letter One embodiment of the device of breath, the Installation practice is corresponding with embodiment of the method shown in Fig. 2, which can specifically answer For in various electronic equipments.
As shown in figure 5, the device 500 for generating information of the present embodiment includes: that acquiring unit 501, first determines list First 502, second determination unit 503, the first generation unit 504.Wherein, acquiring unit 501, for obtaining in prefixed time interval User is directed at least one feedback information that end article is submitted；First determination unit 502, for at least one feedback information Semantic analysis is carried out, determines whether every feedback information at least one feedback information is to be used to indicate end article there are matter The negative feedback information of amount problem；Second determination unit 503, for determining feedback letter negative at least one feedback information The quantity of breath；First generation unit 504, for being fed back according in the quantity of negative feedback information and at least one feedback information The sum of information, for characterizing end article, there are the mass fractions of the probability of quality problems for generation.
In the present embodiment, for generating acquiring unit 501, the first determination unit 502, second of the device 500 of information The specific processing of determination unit 503, the first generation unit 504 can refer to step 201, step in Fig. 2 corresponding embodiment 202, step 203 and step 204.
In some optional implementations of the present embodiment, the first determination unit 502, comprising: generate subelement (in figure not Show), for being pre-processed to every feedback information at least one feedback information, generate corresponding with every feedback information Keyword sequence；First determines subelement (not shown), for determining and every according to obtained keyword sequence The corresponding term vector matrix of feedback information；Subelement (not shown) is imported, for importing identified term vector matrix Trained semantics recognition model in advance obtains the semanteme of every feedback information, and semantics recognition model is for characterizing term vector matrix With the semantic corresponding relationship of feedback information, the semanteme of feedback information includes negative and non-negative.
In some optional implementations of the present embodiment, device further include: third determination unit (not shown) is used In the characteristic information for determining mass fraction according at least one in following item of information: feedback information at least one feedback information Sum, submit the attribute value of the number of users of at least one feedback information, pre-set end article.
In some optional implementations of the present embodiment, the first generation unit 504, comprising: second determines subelement (figure In be not shown), for determining the ratio of the quantity of negative feedback information and the sum of feedback information at least one feedback information Value；Third determines subelement (not shown), for determining mass fraction according to ratio and characteristic information.
In some optional implementations of the present embodiment, the first generation unit 504, comprising: divide subelement (in figure not Show), for the submission time according to every feedback information at least one feedback information, at least one feedback information is drawn It is divided into preset number feedback information set；4th determines subelement (not shown), for determining that preset number is fed back The quantity for the negative feedback information that each feedback information set includes in information aggregate, the sum with the feedback information for including Ratio；First obtains subelement (not shown), is directed to each pre-set weight of feedback information set for obtaining；The Five determine subelement (not shown), for determining quality point according to the corresponding weight of each feedback information set and ratio Number.
In some optional implementations of the present embodiment, device further include: the 4th determination unit (not shown) is used When the delivery availability and feedback information for determining every feedback information is targeted at least one feedback information commodity are submitted Between between time difference；5th determination unit (not shown), for determining that end article is flat according to the identified time difference Feed back lag time.
In some optional implementations of the present embodiment, acquiring unit 501, comprising: inquiry subelement (does not show in figure Out), for inquiring submission time within a preset time interval in historical feedback information aggregate, and believe with the mark of end article Cease matched feedback information；Second obtains subelement (not shown), for obtaining in the feedback information inquired, with inquiry To the associated order information of feedback information and/or the user information not feedback information in pre-set blacklist.
In some optional implementations of the present embodiment, device further include: the second generation unit (not shown) is used In being greater than preset first threshold in response to mass fraction generated, generation is used to indicate end article, and there may be quality to ask The prompt information of topic.
In some optional implementations of the present embodiment, device further include: the 6th determination unit (not shown) is used In determining the difference between mass fraction generated and the mass fraction of a upper time interval；Third generation unit is (in figure not Show), for being greater than preset second threshold in response to difference, generation is used to indicate end article, and there may be quality problems Prompt information.
In some optional implementations of the present embodiment, device further include: the 4th generation unit (not shown) is used Successively decrease in the mass fraction in response to preset number time interval before with mass fraction generated, generation is used to indicate There may be the prompt informations of quality problems for end article.
In some optional implementations of the present embodiment, device further include: sequencing unit (not shown), for pair The mass fraction of mass fraction generated and other commodity in the affiliated classification of end article in addition to end article is ranked up； 5th generation unit (not shown), for according to ranking results determine whether generate be used to indicate end article there may be The prompt information of quality problems.
The device provided by the above embodiment of the application is mentioned by user in acquisition prefixed time interval for end article At least one feedback information handed over then carries out semantic analysis at least one feedback information, determines at least one feedback information In every feedback information whether be the negative feedback information for being used to indicate end article there are quality problems, and determine at least The quantity of negative feedback information in one feedback information is finally fed back according to the quantity of negative feedback information and at least one The sum of feedback information in information, for characterizing end article, there are the mass fractions of the probability of quality problems for generation, to mention A kind of commercial quality monitoring mechanism based on feedback information has been supplied, commercial quality monitoring method is enriched.
Below with reference to Fig. 6, it illustrates the computer systems 600 for the server for being suitable for being used to realize the embodiment of the present application Structural schematic diagram.Server shown in Fig. 6 is only an example, should not function and use scope band to the embodiment of the present application Carry out any restrictions.
As shown in fig. 6, computer system 600 includes central processing unit (CPU) 601, it can be read-only according to being stored in Program in memory (ROM) 602 or be loaded into the program in random access storage device (RAM) 603 from storage section 608 and Execute various movements appropriate and processing.In RAM 603, also it is stored with system 600 and operates required various programs and data. CPU 601, ROM 602 and RAM 603 are connected with each other by bus 604.Input/output (I/O) interface 605 is also connected to always Line 604.
I/O interface 605 is connected to lower component: the importation 606 including keyboard, mouse etc.；It is penetrated including such as cathode The output par, c 607 of spool (CRT), liquid crystal display (LCD) etc. and loudspeaker etc.；Storage section 608 including hard disk etc.； And the communications portion 609 of the network interface card including LAN card, modem etc..Communications portion 609 via such as because The network of spy's net executes communication process.Driver 610 is also connected to I/O interface 606 as needed.Detachable media 611, such as Disk, CD, magneto-optic disk, semiconductor memory etc. are mounted on as needed on driver 610, in order to read from thereon Computer program be mounted into storage section 608 as needed.
Particularly, in accordance with an embodiment of the present disclosure, it may be implemented as computer above with reference to the process of flow chart description Software program.For example, embodiment of the disclosure includes a kind of computer program product comprising be carried on computer-readable medium On computer program, which includes the program code for method shown in execution flow chart.In such reality It applies in example, which can be downloaded and installed from network by communications portion 609, and/or from detachable media 611 are mounted.When the computer program is executed by central processing unit (CPU) 601, limited in execution the present processes Above-mentioned function.It should be noted that computer-readable medium described herein can be computer-readable signal media or Computer readable storage medium either the two any combination.Computer readable storage medium for example can be --- but Be not limited to --- electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor system, device or device, or any above combination. The more specific example of computer readable storage medium can include but is not limited to: have one or more conducting wires electrical connection, Portable computer diskette, hard disk, random access storage device (RAM), read-only memory (ROM), erasable type may be programmed read-only deposit Reservoir (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory Part or above-mentioned any appropriate combination.In this application, computer readable storage medium, which can be, any include or stores The tangible medium of program, the program can be commanded execution system, device or device use or in connection.And In the application, computer-readable signal media may include in a base band or the data as the propagation of carrier wave a part are believed Number, wherein carrying computer-readable program code.The data-signal of this propagation can take various forms, including but not It is limited to electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be computer Any computer-readable medium other than readable storage medium storing program for executing, the computer-readable medium can send, propagate or transmit use In by the use of instruction execution system, device or device or program in connection.Include on computer-readable medium Program code can transmit with any suitable medium, including but not limited to: wireless, electric wire, optical cable, RF etc., Huo Zheshang Any appropriate combination stated.
Flow chart and block diagram in attached drawing are illustrated according to the system of the various embodiments of the application, method and computer journey The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation A part of one unit of table, program segment or code, a part of the unit, program segment or code include one or more Executable instruction for implementing the specified logical function.It should also be noted that in some implementations as replacements, institute in box The function of mark can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are practical On can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it wants It is noted that the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart, Ke Yiyong The dedicated hardware based system of defined functions or operations is executed to realize, or can be referred to specialized hardware and computer The combination of order is realized.
Being described in unit involved in the embodiment of the present application can be realized by way of software, can also be by hard The mode of part is realized.Described unit also can be set in the processor, for example, can be described as: a kind of processor packet Include acquiring unit, the first determination unit, the second determination unit, the first generation unit.Wherein, the title of these units is in certain feelings The restriction to the unit itself is not constituted under condition, for example, acquiring unit is also described as " obtaining in prefixed time interval The unit at least one feedback information that user submits for end article ".
As on the other hand, present invention also provides a kind of nonvolatile computer storage media, the non-volatile calculating Machine storage medium can be nonvolatile computer storage media included in device described in above-described embodiment；It is also possible to Individualism, without the nonvolatile computer storage media in supplying server.Above-mentioned nonvolatile computer storage media It is stored with one or more program, when one or more of programs are executed by an equipment, so that the equipment: obtaining User in prefixed time interval is taken to be directed at least one feedback information that end article is submitted；At least one feedback information is carried out Semantic analysis determines whether every feedback information at least one feedback information is to be used to indicate end article there are quality to ask The negative feedback information of topic；Determine the quantity of feedback information negative at least one feedback information；According to negative feedback The sum of feedback information in the quantity of information and at least one feedback information, for characterizing end article, there are quality problems for generation Probability mass fraction.
Above description is only the preferred embodiment of the application and the explanation to institute's application technology principle.Those skilled in the art Member is it should be appreciated that invention scope involved in the application, however it is not limited to technology made of the specific combination of above-mentioned technical characteristic Scheme, while should also cover in the case where not departing from the inventive concept, it is carried out by above-mentioned technical characteristic or its equivalent feature Any combination and the other technical solutions formed.Such as features described above has similar function with (but being not limited to) disclosed herein Can technical characteristic replaced mutually and the technical solution that is formed.

Claims (24)

1. a kind of method for generating information, which is characterized in that the described method includes:
It obtains user in prefixed time interval and is directed at least one feedback information that end article is submitted；
Semantic analysis is carried out at least one feedback information, determines every feedback letter at least one feedback information Whether breath is the negative feedback information for being used to indicate the end article there are quality problems；
Determine the quantity of feedback information negative at least one feedback information；
According to the sum of feedback information in the quantity of the negative feedback information and at least one feedback information, generates and use In characterizing the end article, there are the mass fractions of the probability of quality problems.
2. the method according to claim 1, wherein described carry out semantic point at least one feedback information Analysis determines whether every feedback information at least one feedback information is to be used to indicate the end article there are quality The negative feedback information of problem, comprising:
Every feedback information at least one feedback information is pre-processed, is generated corresponding with every feedback information Keyword sequence；
According to obtained keyword sequence, term vector matrix corresponding with every feedback information is determined；
Identified term vector matrix is imported to semantics recognition model trained in advance, obtains the semanteme of every feedback information, institute Predicate justice identification model is used to characterize the semantic corresponding relationship of term vector matrix and feedback information, and the semanteme of feedback information includes It is negative and non-negative.
3. the method according to claim 1, wherein the method also includes:
The characteristic information of the mass fraction: at least one feedback information is determined according at least one in following item of information The sum of middle feedback information, the number of users for submitting at least one feedback information, the pre-set end article category Property value.
4. according to the method described in claim 3, it is characterized in that, the quantity and institute according to the negative feedback information The sum of feedback information at least one feedback information is stated, there are the probability of quality problems for characterizing the end article for generation Mass fraction, comprising:
Determine the ratio of the sum of feedback information in the quantity and at least one feedback information of the negative feedback information；
According to the ratio and the characteristic information, the mass fraction is determined.
5. the method according to claim 1, wherein the quantity and institute according to the negative feedback information The sum of feedback information at least one feedback information is stated, there are the probability of quality problems for characterizing the end article for generation Mass fraction, comprising:
According to the submission time of every feedback information at least one feedback information, by least one feedback information It is divided into preset number feedback information set；
Determine the number for the negative feedback information that each feedback information set includes in the preset number feedback information set Amount, the ratio with the sum for the feedback information for including；
It obtains and is directed to each pre-set weight of feedback information set；
According to the corresponding weight of each feedback information set and ratio, the mass fraction is determined.
6. the method according to claim 1, wherein the method also includes:
Determine that the delivery availability for the commodity that every feedback information is targeted at least one feedback information is mentioned with feedback information Time difference between the time of friendship；
End article average feedback lag time is determined according to the identified time difference.
7. the method according to claim 1, wherein user is directed to target quotient in the acquisition prefixed time interval At least one feedback information that product are submitted, comprising:
Submission time is inquired in historical feedback information aggregate in the prefixed time interval, and the mark with the end article Know the feedback information of information matches；
It obtains in the feedback information inquired, does not exist with the associated order information of feedback information and/or user information inquired Feedback information in pre-set blacklist.
8. method according to any one of claims 1-7, which is characterized in that the method also includes:
In response to mass fraction generated be greater than preset first threshold, generation be used to indicate the end article there may be The prompt information of quality problems.
9. method according to any one of claims 1-7, which is characterized in that the method also includes:
Determine the difference between mass fraction generated and the mass fraction of a upper time interval；
It is greater than preset second threshold in response to the difference, generation is used to indicate the end article, and there may be quality problems Prompt information.
10. method according to any one of claims 1-7, which is characterized in that the method also includes:
Successively decrease in response to the mass fraction and mass fraction generated of preset number time interval before, generates for referring to Showing the end article, there may be the prompt informations of quality problems.
11. method according to any one of claims 1-7, which is characterized in that the method also includes:
To the matter of other commodity in mass fraction generated and the affiliated classification of the end article in addition to the end article Amount score is ranked up；
Determine whether to generate according to ranking results and is used to indicate the end article there may be the prompt informations of quality problems.
12. a kind of for generating the device of information, which is characterized in that described device includes:
Acquiring unit is directed at least one feedback information that end article is submitted for obtaining user in prefixed time interval；
First determination unit determines at least one feedback for carrying out semantic analysis at least one feedback information Whether every feedback information in information is the negative feedback information for being used to indicate the end article there are quality problems；
Second determination unit, for determining the quantity of feedback information negative at least one feedback information；
First generation unit, for being fed back according in the quantity of the negative feedback information and at least one feedback information The sum of information, for characterizing the end article, there are the mass fractions of the probability of quality problems for generation.
13. device according to claim 12, which is characterized in that first determination unit, comprising:
Subelement is generated, for being pre-processed to every feedback information at least one feedback information, is generated and every The corresponding keyword sequence of feedback information；
First determines subelement, for determining term vector corresponding with every feedback information according to obtained keyword sequence Matrix；
Subelement is imported, for identified term vector matrix to be imported to semantics recognition model trained in advance, obtains every instead The semanteme of feedforward information, the semantics recognition model are used to characterize the semantic corresponding relationship of term vector matrix and feedback information, instead The semanteme of feedforward information includes negative and non-negative.
14. device according to claim 12, which is characterized in that described device further include:
Third determination unit, for determining the characteristic information of the mass fraction: institute according at least one in following item of information It states the sum of feedback information at least one feedback information, the number of users for submitting at least one feedback information, preset The end article attribute value.
15. device according to claim 14, which is characterized in that first generation unit, comprising:
Second determines subelement, anti-in the quantity and at least one feedback information for determining the negative feedback information The ratio of the sum of feedforward information；
Third determines subelement, for determining the mass fraction according to the ratio and the characteristic information.
16. device according to claim 12, which is characterized in that first generation unit, comprising:
Subelement is divided, it, will be described for the submission time according to every feedback information at least one feedback information At least one feedback information is divided into preset number feedback information set；
4th determines subelement, and for determining, each feedback information set includes in the preset number feedback information set The quantity of negative feedback information, the ratio with the sum for the feedback information for including；
First obtains subelement, is directed to each pre-set weight of feedback information set for obtaining；
5th determines subelement, for determining the mass fraction according to the corresponding weight of each feedback information set and ratio.
17. device according to claim 12, which is characterized in that described device further include:
4th determination unit, for determining the delivery of every feedback information is targeted at least one feedback information commodity The time difference between time that time and feedback information are submitted；
5th determination unit, for determining end article average feedback lag time according to the identified time difference.
18. device according to claim 12, which is characterized in that the acquiring unit, comprising:
Subelement is inquired, for inquiring submission time in historical feedback information aggregate in the prefixed time interval, and with The feedback information of the identification information match of the end article；
Second obtains subelement, for obtaining in the feedback information inquired, believes with the associated order of feedback information inquired Breath and/or the user information not feedback information in pre-set blacklist.
19. device described in any one of 2-18 according to claim 1, which is characterized in that described device further include:
Second generation unit, for being greater than preset first threshold in response to mass fraction generated, generation is used to indicate institute State prompt information of the end article there may be quality problems.
20. device described in any one of 2-18 according to claim 1, which is characterized in that described device further include:
6th determination unit, for determining the difference between mass fraction generated and the mass fraction of a upper time interval；
Third generation unit, for being greater than preset second threshold in response to the difference, generation is used to indicate the target quotient There may be the prompt informations of quality problems for product.
21. device described in any one of 2-18 according to claim 1, which is characterized in that described device further include:
4th generation unit, for the mass fraction and quality generated point in response to preset number time interval before Number successively decreases, and generation is used to indicate the end article, and there may be the prompt informations of quality problems.
22. device described in any one of 2-18 according to claim 1, which is characterized in that described device further include:
Sequencing unit, for in mass fraction generated and the affiliated classification of the end article in addition to the end article The mass fraction of other commodity is ranked up；
5th generation unit is used to indicate the end article there may be quality for determining whether to generate according to ranking results The prompt information of problem.
23. a kind of server, comprising:
One or more processors；
Storage device, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processors Realize the method as described in any in claim 1-11.
24. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is by processor The method as described in any in claim 1-11 is realized when execution.
CN201710758012.4A 2017-08-29 2017-08-29 Method and apparatus for generating information Pending CN109426978A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710758012.4A CN109426978A (en) 2017-08-29 2017-08-29 Method and apparatus for generating information

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710758012.4A CN109426978A (en) 2017-08-29 2017-08-29 Method and apparatus for generating information

Publications (1)

Publication Number Publication Date
CN109426978A true CN109426978A (en) 2019-03-05

Family

ID=65503597

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710758012.4A Pending CN109426978A (en) 2017-08-29 2017-08-29 Method and apparatus for generating information

Country Status (1)

Country Link
CN (1) CN109426978A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115512529A (en) * 2021-06-23 2022-12-23 中国石油化工股份有限公司 Contractor problem early warning method, early warning device and early warning system

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110173191A1 (en) * 2010-01-14 2011-07-14 Microsoft Corporation Assessing quality of user reviews
CN103123633A (en) * 2011-11-21 2013-05-29 阿里巴巴集团控股有限公司 Generation method of evaluation parameters and information searching method based on evaluation parameters
CN103577988A (en) * 2012-07-24 2014-02-12 阿里巴巴集团控股有限公司 Method and device for recognizing specific user
CN103839178A (en) * 2014-02-28 2014-06-04 珠海市君天电子科技有限公司 Method and system for obtaining commodity quality information
CN106157119A (en) * 2016-07-11 2016-11-23 广东聚联电子商务股份有限公司 A kind of method that e-commerce purchases system platform report processes
CN106846015A (en) * 2017-02-06 2017-06-13 吉林师范大学 Product quality reponse system and application method based on Internet of Things

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110173191A1 (en) * 2010-01-14 2011-07-14 Microsoft Corporation Assessing quality of user reviews
CN103123633A (en) * 2011-11-21 2013-05-29 阿里巴巴集团控股有限公司 Generation method of evaluation parameters and information searching method based on evaluation parameters
CN103577988A (en) * 2012-07-24 2014-02-12 阿里巴巴集团控股有限公司 Method and device for recognizing specific user
CN103839178A (en) * 2014-02-28 2014-06-04 珠海市君天电子科技有限公司 Method and system for obtaining commodity quality information
CN106157119A (en) * 2016-07-11 2016-11-23 广东聚联电子商务股份有限公司 A kind of method that e-commerce purchases system platform report processes
CN106846015A (en) * 2017-02-06 2017-06-13 吉林师范大学 Product quality reponse system and application method based on Internet of Things

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115512529A (en) * 2021-06-23 2022-12-23 中国石油化工股份有限公司 Contractor problem early warning method, early warning device and early warning system
CN115512529B (en) * 2021-06-23 2024-03-05 中国石油化工股份有限公司 Contractor problem early warning method, early warning device and early warning system

Similar Documents

Publication Publication Date Title
CN107908740B (en) Information output method and device
CN107391680A (en) Content recommendation method, device and equipment
US11127032B2 (en) Optimizing and predicting campaign attributes
CN109840730B (en) Method and device for data prediction
US11200591B2 (en) Electronic content based on neural networks
CN110019699A (en) Pass through the classification of grammer slot between domain
CN109976997A (en) Test method and device
CN110929136A (en) Personalized recommendation method and device
CN108921323A (en) Method and apparatus for generating information
CN110363604A (en) Page generation method and device
CN111400613A (en) Article recommendation method, device, medium and computer equipment
CN109544076A (en) Method and apparatus for generating information
JP2019032620A (en) Generation device, method for generation, and generation program
WO2020211616A1 (en) Method and device for processing user interaction information
CN113743971A (en) Data processing method and device
US10628806B2 (en) System and method for test data provisioning
US20170249697A1 (en) System and method for machine learning based line assignment
CN109426978A (en) Method and apparatus for generating information
CN107357847B (en) Data processing method and device
CN116186541A (en) Training method and device for recommendation model
CN113435967B (en) Method and system for automatically determining marketing content of agricultural products
US20210350202A1 (en) Methods and systems of automatic creation of user personas
Hou et al. Pricing leadership decisions of competing firms with consumer learning
Briones-Castañeda et al. Digital transformation model with a focus on total quality management and lean manufacturing to increase online sales in textile SMEs
JP2019032827A (en) Generation device, method for generation, and generation program

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20190305

RJ01 Rejection of invention patent application after publication